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 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
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
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.
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.
It's worse than that.
Many people who created extensions have given up and abandoned them. Its not that they don't want to re-write them, its because their extensions CAN'T be re-written to work with the new Firefox due to the internal changes that have been made.
And that leads to the real problem -- Firefox has always been shit, but, we were able to cover up all of its flaws and turn it into a good browser with extensions. Now that all my favorite themes and extensions no longer work, I'm just left with a shit browser that no longer works or looks the way I want.
Fortunately, there's still Palemoon.
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
HTML5 is like C++. A massive, hard to implement standard that barely anyone fully understands, let alone uses. Supporting about 80% of it gives you 99% compatibility, so it's hard to justify the massive effort required to finish the last, really difficult 20% off.
const int one = 65536; (Silvermoon, Texture.cs)
SJW, n: "Someone I don't like, and by the way I'm a fuckwit" - AC
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.