Slashdot Mirror
Firefox 58 Gets Graphics Speed Boost, Web App Abilities (cnet.com)
Mozilla released on Tuesday a new version of its Firefox Quantum browser, boosting its graphics speed and improving a couple of new technologies designed to make the web more powerful. From a report: The browser, version 58, is the first major update since Mozilla's recovery plan hit full stride in November with the debut of Firefox Quantum. Speed is of the essence in Mozilla's recovery plan, and Firefox 58 does better than its predecessor in some graphics tasks by splitting work better across the multiple processor cores that computer chips have these days. The result should be scrolling that's smooth, uninterrupted by the stuttering that in computing circles goes by the disparaging term "jank." [...] Firefox 58 helps with two new web technologies. One, called WebAssembly, provides for dramatically faster web apps. Firefox 58 can get WebAssembly software running faster so you don't have to twiddle your thumbs waiting as long after clicking a link. Another is progressive web apps (PWAs), an initiative that came out of Google to help make the web a better match for the apps we all drop on our phones.
I'm very grateful to the Firefox project and its contributors for their dedication to bringing us a fast and modern browser to act not only as a useful product, but as an essential counterweight to corporate hegemony over the www. Switching to 57 was a bit of pain as I had to find replacements to many of my beloved extensions, but it was worth it for the speed upgrade and smaller memory footprint. I'm glad they are keeping on the path of optimization and bringing more technologies that I can use both as an end user and as a web developer.
I doubt the fixed it, but the new Quantum "faster" Firefox was really dragging down my system. At first, I thought some malicious add-on was mining cryptocurrency on my machine. But it turns out Firefox was just spawning orphan processes. I found the fix at the link below, which is basically to disable multi-threading in Firefox.
Multiple Firefoxes in the background, exiting the program doesn't clear them up. They persist.
I am still missing a few of my favorite add-ons as well. The bulk download manager DownThemAll was great, but it sounds like Firefox does not want that functionality, so no add-on has yet to be as useful.
I think there's a good amount of "written before me" attitudes causing problems.
Maintaining the code of someone else is seldom attractive to developers, who would rather make their own mark, and refuse to entertain the idea that what they create might be worse than what was already there.
Re-inventing the wheel seldom leads to an improvement on the circular shape, centered hub and perpendicular axis.
Focusing on improving the core technology is the right decision IMHO. Recent performance improvements have been quite impressive, and the distance to Chrome has become really small.
Nevertheless I would really like to see a way to measure webworker performance. Sometimes I have the feeling that there is quite some fluctuation. For example when I work with iconfu.com, sometimes the icons get rendered blazingly fast, and sometimes it takes seconds. Not sure what is causing this, also since I cannot measure webworker performance, there is not really an easy way to find out.
Anyway, keep up the great work!
Signature deleted by lameness filter.
... if Firefox put back some of the lost functionality due to all the extensions that no longer work due to Mozilla's apparent race to be a Chrome clone.
The only thing that separated Firefox from the competition were the plugins (to be more exact, the power of these plugins) and Firefox threw it out and replaced it with lobotomized versions incapable to replicate the functionality of the previous versions.
So I will stay with Firefox 56 (the last sane version) as much as I can.
Religion: The greatest weapon of mass destruction of all time
Chrome nor Firefox use acceleration on linux using intel graphics
Be or ben't
We've had portable binary formats on the web for years - originally java bytecode and then flash. And look how bug and exploit free thar was. Why the rush to produce yet another attack surface?
Here's how Mozilla lost its 1st place in one of their last remaining bastions. "Speed is of the essence in Mozilla's recovery plan" the article says, but in reality what should be the essence for a recovery plan is to bite the bullet and admit they were wrong in deprecating what made them unique.
I'll stick with Waterfox. What made Firefox so great was the ability to bend it to my will. Webextensions takes away my abilities and replaces them with training wheels and a balloon. YAY! This update is just more hand waving "Pay no attention to the man behind the curtain" tactics.
"A person is smart. People are dumb, panicky dangerous animals and you know it." - K
I am still not switching back until all the old add ons work.
The millennial that doesn't like most of the stuff designed for millennials.
While I'm all for Firefox improving performance, it can't come at the expensive of reliably rendering websites. After the upgrade, I noticed various problems across several sites I used to access without issue.
Because of that, I've been forced to switch to Chrome for the time being as my primary browser because I have enough on my plate without having to worry about "Is the website broken or is it me?"
Maybe I'll try it again after it's had a couple versions to shake out bugs.
Did. They suck. Made me itchy.
The millennial that doesn't like most of the stuff designed for millennials.
Palemoon / Waterfox is maintained by 1 person... ONE; and at most, 2.
Yes, there's been minor chip-ins from a few others, but look at the git commit history and it's basically a one-man show.
And seeing how fast tech is moving and standards are evolving, Palemoon/Waterfox/Variants are already far, far behind, especially in security fixes.
It's basically a dead project without some serious number of hands contributing to maintaining it! And that of course requires serious funding, or at least, some corporate sponsorship with lots of devs.
Web browsers are probably some of the most complex pieces of applications, perhaps more so than even Operating Systems, so it's unrealistic for any 1 person to maintain, let alone develop further.
On Windows 7
Firefox v58 Gets an HTML5 Test score of 486 out of 555
Chrome v63 Gets an HTML5 Test score of 528 out of 555
I am in general still perplex why after all these years browsers are not 100% HTML 5 complaint.
If something is so important that you feel the need to post it on the internet... It probably isn't that important.
Fuck off.
grandpa REALLY needs new diapers
Off topic, but I believe relevant to what you wrote. I also noticed in one of my desktops, using Windows 10, that when an old game I have tried to use near 4GB of RAM Windows simply stopped responding, and this considering that the desktop in question have 8GB of RAM (It should be more than enough for the game to work). I also wanted to know what they did with memory management, since the same thing does not happen on the other simmilar desktop with Windows 7 (the game works correctly or if it does not work it don't crash the operating system)
Religion: The greatest weapon of mass destruction of all time
Re-inventing the wheel seldom leads to an improvement on the circular shape, centered hub and perpendicular axis.
Mozilla's new wheel is a triangle, attached to a bald feminist.
Nope. I like the old brand.
The millennial that doesn't like most of the stuff designed for millennials.
In the run up to FF57's deadline came up last year, I bitterly posted on Slashdot about how I didn't want the speed upgrades as much as I wanted to keep the extensions that were not getting ported.
I was wrong - dead wrong. Why?
1) Speed: If you were an anti-Chrome guy like me but would be a little jealous of its speed when you had to use it, this has been resolved. FF57 has been much snappier to use than previous versions. It feels like Chrome or faster.
In the end, browser speed DOES matter.
2) Extensions: Not every extension I used before FF57 has been ported to Quantum, but the important extensions I used have been since the FF57 release that weren't ready initially. NoScript, FlagFox, etc. All working now. If the extensions weren't ready last November, look again. They may be ready now.
Just as importantly, I haven't missed the ones that haven't made it. You may not miss them either.
3) The native Web Development tools are better than Firebug was in the end, IMO. And it's been nice to not have Firebugginess to deal with anymore.
So - Take it from this OCD guy who's not a marketing shill for the Mozilla org: If you're still sitting at FF52 ESR or FF56.x over any form of FUD (especially the, "but extension X is indispensible", worry,) given FF57 another look now and test it out without those, "I've gotta have extension X," extensions and see. You may be surprised.
While people everywhere are telling us why we CAN'T do something we are out there proving them wrong. Alex isn't working alone he is simply taking the best parts of Firefox and stripping out the junk. Most of the improvements Quantum made are already incorporated into Waterfox. It just didn't implement the nanny state of removing things that many users find indispensable. Do you know how many Fortune 500 companies have JAVA front ends for their multi-million dollar enterprises? A LOT! You know what most of these companies are using for a browser? IE11!!! That is SCARY. At least with Waterfox you've got a modern browser that still allows legacy sites to work correctly.
"A person is smart. People are dumb, panicky dangerous animals and you know it." - K
Time to talk about how Firefox/Chrome/etc are all bad and dumb and only I know how to make the one true browser that would make everyone happy!
And that's one of the problems with browser development culture - "if you lag behind, we don't care about you. Doesn't matter if your use case is completely valid, you're in the minority and you don't matter." That's the feeling I get.
Perhaps you don't realise the main reason corporate environments are using IE is because it's bundled with Windows, available on every corporate machine; and best of all: it can be remotely configured / controlled / patched using domain controller / active directory rules and WSUS. And if things go wrong, they have license agreements with Microsoft, meaning dedicated support and swift responses.
I recommend that you read Firefox's privacy policy. Its "privacy controls", as you put it, are quite suspect.
The Firefox privacy policy dated September 28, 2017 clearly indicates that it can/will send data to Mozilla, along with third parties like Google, Adjust, SalesForce, and Leanplum:
It doesn't matter if such data collection and transmission can "potentially" be disabled.
If Firefox's developers really gave a damn about Firefox's users' privacy, then Firefox wouldn't even include any support for any of that tracking and sending of private data to third parties. There'd be nothing to disable, as the code implementing such tracking and data transmission shouldn't even exist in Firefox!
The Rust programming language is a great example of how Mozilla's developers are going out of their way to create a shitty imitation of something that already exists and does the job much better.
The Rust home page describes Rust like this:
At a glance, all of that sounds nice to have. But when you give Rust a try, I think it soon becomes obvious that you're in for a world of hurt.
Rust's approach to memory management is perhaps the worst I've ever had to deal with in my decades of experience. Even when you have a full understanding of it, it's still awkward to use. While C's approach might be somewhat dangerous, at least it allows a professional to get real work done efficiently. When using Rust, it's like you have to be an expert user just to get anything accomplished.
I don't want to get into it in too much detail, but Rust suffers from a number of what I think are some very serious problems, including a mediocre imitation of a C-like syntax, a rather lacking standard library, a single implementation that's quite slow, often broken/incomplete third party libraries, far too much hype, and a rather toxic/tyrannical community.
The thing that's really devastating to Rust, however, is C++. Modern versions of C++, like C++14 and especially C++17, basically remove the alleged benefits of Rust, while at the same time offering many benefits that Rust doesn't, and may never, offer. It's hard to believe, but modern versions of C++ probably have an easier learning curve than Rust has. There are also multiple independent C++ compiler and standard library implementations, supporting many platforms. There are many high-quality, stable, tested C++ libraries available. C++ has one of the nicest, most friendly and helpful communities around.
Instead of doing the sensible thing and pushing hard to use all of the beneficial features of C++14 and C++17 while developing Firefox, we've seen Mozilla's developers waste a lot of resources building Rust. All they've ended up with in the end is a difficult to use language that's inferior to C++17 in almost every way.
I think that this failing Rust effort, along with other failures like Firefox OS and Australis, are all big reasons why Firefox has declined so severely over the last number of years. Instead of just listening to Firefox's users and giving them what they want (i.e. a fast, extensible, secure browser), we've seen Firefox's developers dick around with so much stuff that Firefox's users don't want, or don't care about.
I believe that terminating this failing Rust effort immediately would be one of the best things that Mozilla and Firefox's developers could do. Stripping out the Rust code that has made its way into Firefox should also become a priority. Firefox needs to focus on using C++17 instead of Rust, in my opinion.
To be fair I think they got the performance improvements, the problem is they broke everything else in the process.
Religion: The greatest weapon of mass destruction of all time
Perhaps you don't realize that the reason they are using IE is they have legacy systems running ActiveX/JAVA programs that only work with IE. Most web sites don't even test against IE any longer they just pop up a message saying "Download Chrome Now!". Which means these companies have to adopt a two browser strategy to support their needs.
"A person is smart. People are dumb, panicky dangerous animals and you know it." - K
Honest question: Waterfox have the performance improvements from 57~beyond AND the support to XUL-style addons from 56 and later?
Religion: The greatest weapon of mass destruction of all time
See if your extensions work under Firefox ESR, it's at version 52.6.0. This version is supposed to include updates to mitigate Spectre. I'm using it now, none of the extensions I use have had problems.
For the most part yes. From the Waterfox site:
Features Disabled Encrypted Media Extensions (EME)
Disabled Web Runtime (deprecated as of 2015)
Removed Pocket
Removed Telemetry
Removed data collection
Removed startup profiling
Allow running of all 64-Bit NPAPI plugins
Allow running of unsigned extensions
Removal of Sponsored Tiles on New Tab Page
Addition of Duplicate Tab option
Locale selector in about:preferences > General
And from the developer Alex Kontos:
"Quantum is already a part of Waterfox. It was just a collection of improvements in various aspects of the browser made at the same time, so Mozilla just decided to use the umbrella term Quantum so people would know that all these changes were happening. It has been going on for years, and Firefox 52+ were the finishing touches to it."
"A person is smart. People are dumb, panicky dangerous animals and you know it." - K
You can always filter ACs. Or have fun.
Waterfox have the performance improvements from 57~beyond
No. Waterfox is still based on Firefox 56. Waterfox will just keep falling further behind.
Yet your shitty company cannot make a browser that is as good as that one guy.
Go shill somewhere else.
Firefox 52+ were the finishing touches to it.
No, there have been further performance improvements since 52. Try this simple example of the improvement in WebAssembly complication times.
In WaterFox 56.0.3 the highest result I got was: WebAssembly.instantiate took 1369.3 ms (9 MB/s)
In Firefox 58.0 the lowest result I got was: WebAssembly.instantiate took 222.5 ms (55.6 MB/s)
Waterfox will continue to fall behind as new Firefox releases come out. Eventually Waterfox will have to bite the bullet and rebase on whatever the latest Firefox is at that time.
The strongest argument against rust is portability. In order to get firefox to build, you first have to port a specific LLVM version, then rust, then firefox. It's an insane amount of work just to get a browser to compile.
In the good old days netscape and later firefox ran on a slew of platforms. You could run it on OS/2, windows, mac, linux, *bsd, solaris, hp ux, etc. Today, Mozilla only cares about the big 3 (win, mac, linux). Google is doing the same. That means we no longer have a true open source, cross platform web browser that's any good.
MidnightBSD: The BSD for Everyone
clearly indicates that it can/will send data to Mozilla
So turn telemetry off in the settings. It's under Privacy and Security -> Firefox Data Collection and Use. While you're there you also might want to set the Tracking Protection setting to "always". That's what I do.
Mommy handjobs are the best
The millennial that doesn't like most of the stuff designed for millennials.
A short while ago I noticed Firefox's memory use is mainly increasing due to sites with certain Java Script, once you disable JS the memory use remains stable.
"The likes of Facebook and WhatsApp are free to those whose privacy is of zero value."
I see you're bitter about the loss of XUL, but every other browser's add-ons are built on chrome's web extension system at this point. Firefox is still the only exception to this since they are adding features that Google refuses to add to chrome despite heavy lobbying to do so on the part of many developers and users.
So far, that I'm aware of, Firefox has added support for noscript (even though it's popular, Google has refused to add support for it) and support for cookie autodelete to be able to delete localstorage data. No browser other than Firefox can provide that level of granular control over it, and if you're privacy conscious, that's a pretty important feature to have.
Given that Waterfox 56.0.3 was released a few weeks ago and Firefox 58 was just released I'll wait for 56.0.4 or later to do an apples to useful browser comparison. I DID however do a WF 56.0.3 to FF 57.0.4 comparison and they are pretty much identical with WF just edging out FF. I'm not saying you're wrong but your test is certainly unfair to Alex and Waterfox.
Waterfox 56.0.3
WebAssembly.instantiate took 1192.2 ms (10.4 MB/s)
Firefox 57.0.4
WebAssembly.instantiate took 1197.8 ms (10.3 MB/s)
"A person is smart. People are dumb, panicky dangerous animals and you know it." - K
but your test is certainly unfair to Alex and Waterfox
Nope, perfectly fair. I tested current release against current release. Waterfox 56.0.3 was released on the 12th of January.
The way it looks right now is that we'll be using it forever. There's no timeline for when TabMixPlus will be ready for the new Firefox. When, or if, it is finally ready, I'll take another look at Firefox 57 or 58.
I also like having the status bar, and I see no way to do it in the new Firefox.
Sometimes I think I'd be willing to PAY for a good browser that works right and respects my privacy.
Merely disabling these privacy-invading, malware-like aspects of Firefox is not sufficient.
Yes it is. Disable it and be happy.
Don't complain if a browser is not compliant with HTML5 ;-)
HTML5 IS A LIVING STANDARD. There never will be HTML 6, the spec is constantly changing; it is alive. This was decided because of the history of compliance and the organic nature of new features being added between new standards which took a long time to formalize or ended up in many small sub-standards.
Some HTML5 standards begin as browser projects for new features which end up being accepted into the standard. The originating browser generally has a leap ahead on new feature additions; plus it helps Chrome that Google has some people working deep within the spec... Mozilla has been busy revamping their browser and coping with bad management which has put them behind (or working on new feature tangents.)
As far as Test scores... depends upon who is writing the tests. A solid set of tests on a feature would be better since Microsoft has a history of testing positive for a feature they do not properly implement. I would not be surprised if Edge tests highly on a widespread popular test set because they would tend to work towards that, given their prior history.
HTML5 draft was getting books and browser claims for support that were poor... and misleading. At draft, features were dropped and changed - I was there. Making it a living standard embraces how things actually operate in the real world.
Democracy Now! - uncensored, anti-establishment news
Waterfox will continue to fall behind as new Firefox releases come out. Eventually Waterfox will have to bite the bullet and rebase on whatever the latest Firefox is at that time.
What an ass. Your statement has been challenged so you resort to juvenile nana nana boo boo tactics. You know damn well Waterfox will release an update based on this release. Whether it will perform better or worse remains to be seen. But God forbid someone suggest you might be wrong.
"A person is smart. People are dumb, panicky dangerous animals and you know it." - K
You know damn well Waterfox will release an update based on this release.
When? Unsurprisingly, a single developer is outperformed by a team of developers. Waterfox is now two releases behind Firefox.
But God forbid someone suggest you might be wrong.
I'm not wrong. Firefox 58 massively outperforms Waterfox 56.0.3 in WebAssembly compilation.
Well, but now it is dying a slow death (see the market share) because Mozilla broke all that addons and customizations which were the reason to use firefox instead of other browsers in the first place... That does not make much sense right?
Religion: The greatest weapon of mass destruction of all time
It's really not. The data is always collected even if it's not transmitted so they can backdoor into it via "experiments" which send all sorts of data back regardless of your privacy setting. This includes personally identifiable information as it's exempted from their normal privacy rules.
The data is always collected even if it's not transmitted
Cool, where's it stored? These content-free claims are boring. If you've got claims to make back them with evidence.
From the horse's mouth: https://bugzilla.mozilla.org/s...
The question is, why is Firefox continuing to collect telemetry data when explicitly told not to?
The frontend calls Services.telemetry APIs unconditionally, but they won't send data if you've opted out.