News for hackers. Stuff that matters.
Re:What’s Next for Multi-Process Firefox by eagsalazar2 on Tuesday August 02, 2016 @08:25PM (#12214524)
Firefox is always fighting an uphill battle against IE and Chrome who have major advantages in both resources and integration with their respective platforms.

Firefox thrived and grew in an environment where competitors where terrible. They lost the lead to Chrome and unless google really drops the ball I don't see how they can get it back by just being better. They'd have to dramatically leapfrog the competition in a very compelling way and since both MS and Google are investing heavily in their browsers, it isn't clear to me how they could ever do that.

Re:What’s Next for Multi-Process Firefox by eru on Tuesday August 02, 2016 @10:25PM (#12215149)
Actually, on Android Firefox works much better than Chrome exactly because they are independent.

Mobile Firefox allows extension, crucially adblockers. I don't think mobile Chrome will allow those anytime soon.

Re:What’s Next for Multi-Process Firefox by Twirrim on Wednesday August 03, 2016 @12:11AM (#12215532)
The original release was pretty bad, and I ignored it from then on until a couple of weeks ago when I gave it another shot. It's so far superior to Chrome on Android, even without extensions like Adblock. That extension really does add value though, especially with the reduction in power demands.
Re:What’s Next for Multi-Process Firefox by tracker1 on Wednesday August 03, 2016 @02:50AM (#12216110)
Agreed, running so much better than a few months ago when I last tried... Lastpass can now at least see the website I'm on, if they can get form filling worked out, I'd be very happy. uBlock origin seems to be working well now too...
Re:What’s Next for Multi-Process Firefox by tracker1 on Wednesday August 03, 2016 @08:27PM (#12222541)
grr... can't click into a textarea to edit existing text... :-( back to chrome.
Re:What’s Next for Multi-Process Firefox by flukus on Wednesday August 03, 2016 @01:09AM (#12215738)
The original release was the only thing that turned my potato tablet into something somewhat useful.
Re:What’s Next for Multi-Process Firefox by tracker1 on Wednesday August 03, 2016 @02:48AM (#12216102)
It looks to be working MUCH better for me, just tried it again, as the last time I tried it (just a couple months ago) it was slow as molasses, and didn't work with lastpass, now it seems to be working with it, which I appreciate to say the least (though it still won't fill the form, may try the FF plugin directly if there is one).

It also seems to have cleaned up their tabbed interface, closer to current chome (though I actually preferred having chrome tabs with the app views). May also give the current Opera another look. All in all, much improved even in just the past 3-4 months or so since I last tried it.

Re:What’s Next for Multi-Process Firefox by j45 on Wednesday August 03, 2016 @02:03AM (#12215917)
Used Firefox for a very long time on Android and I can agree to this.

Opera Mobile is another tech that I'm enjoying using more and more.

Re:What’s Next for Multi-Process Firefox by tomc1985 on Tuesday August 02, 2016 @11:14PM (#12215315)
It could dramatically leapfrog the competition by returning to its roots -- modularity, power, extensibility... something for the power users.

Instead Moz executives seem set on forging all sorts of weird new business partnerships, and matching Chrome/Safari feature-for-feature. So FF becomes the sad little me-too browser, users everywhere wondering just how long their addons will continue to work...

Re:What’s Next for Multi-Process Firefox by pbuzbee on Wednesday August 03, 2016 @12:03AM (#12215502)
Agreed. Firefox falls short when it competes directly with the better-backed browsers. Its niche (and current advantage) really relies in working well for power users and letting the community find new innovative ways to enhance a browser.
Re:What’s Next for Multi-Process Firefox by rpgmaker on Wednesday August 03, 2016 @10:07AM (#12217964)
It's so bad that they will actually push their loyal users to Chrome. The chromeization of Firefox has been awful and it seems like it's going to continue.
Re:What’s Next for Multi-Process Firefox by Illniyar on Wednesday August 03, 2016 @03:04AM (#12216162)
Well, I think Servo is their leapfrog but in a very different way.

Both IE/Edge and Chrome (and the current Firefox) are based on really old tech stacks with years of legacy code and design decisions. Servo, if it pans out, will be the only "real" modern browser, which should provide a boost in performance and experience that it will take Chrome and IE/Edge years to match.

It would also make iterating on the browser much faster, which could slowly erode Chrome and IE/Edge's lead.

If Firefox can capitalize and successfully market their new browser it could mean becoming the lead browser for years.

Re:What’s Next for Multi-Process Firefox by wooptoo on Wednesday August 03, 2016 @03:15AM (#12216213)
> it will take Chrome and IE/Edge years to match.

Or just a few months to fork and improve. I'd really like to see the Chrome and Firefox teams converge over Servo. But maybe Google suffers from too much NIH syndrome to do that at this point.

Re:What’s Next for Multi-Process Firefox by Sylos on Wednesday August 03, 2016 @06:16AM (#12216913)
So, I was wondering about licensing, if this would mean that Google/Microsoft would have to open-source their browsers in order to use Servo, but seems like Servo's license (MPL) does permit linking to it from code which is under a different license.

The MPL is similar to the LGPL in that way, and KHTML is licensed under LGPL, which is what WebKit was forked from, so should be legally settled at this point, too.

I'm not a lawyer, though, so no guarantees that this is correct.

Re:What’s Next for Multi-Process Firefox by garaetjjte on Wednesday August 03, 2016 @04:26AM (#12216482)
Servo is written in Rust. That isn't going to improve performance.
Re:What’s Next for Multi-Process Firefox by Sylos on Wednesday August 03, 2016 @04:47AM (#12216573)
It should improve performance in the sense that Rust helps to produce code which can be easily parallelized.

But Servo being written in Rust isn't the argument that he's making.

Servo is built with a modern architecture, which all current browsers don't have, because they all started out somewhen pre-2000, before multi-core processors were really a thing, and before the web was really graphical.

What Servo can improve is mainly, that the rendering of HTML itself is parallelized in Servo, and that with WebRender, they utilize graphics cards quite similar to the way that videogames do it, which both do result in noticeable performance improvements.

Re:What’s Next for Multi-Process Firefox by Illniyar on Wednesday August 03, 2016 @05:09AM (#12216675)
Your comment implictly assumes that the programming langauge has a major impact on browser performance.

My expierence with other product development tells me otherwise, and I trust that the people behind mozilla who are specifically building a more performant browser using rust are aware of what does or does not greatly affects performance.

Re:What’s Next for Multi-Process Firefox by yoklov on Wednesday August 03, 2016 @05:20AM (#12216717)
That's not the only difference, or even an important one for performance. Servo's WebRender engine gets 60fps (or in the hundreds if uncapped) on many pages where Chrome, Firefox, Edge, etc. all struggle in the 5-15fps range.

(Of course, Servo is still a long way from production ready).

Re:What’s Next for Multi-Process Firefox by ryuuchin on Tuesday August 02, 2016 @10:41PM (#12215200)
Part of me thinks it's a lack of direction which has limited Mozilla/Firefox in the past/currently(?). Like the dropping of 64-bit support (which they reversed) which I think showed a clear lack of direction on what they wanted to do. Granted this was a few years ago now but it was still clear at that point that 64-bit provided a number of useful (security) features over 32-bit software.

Maybe I'm wrong but that's just how it appears as someone looking in from the outside and who doesn't use Firefox on a daily basis.

Re:What’s Next for Multi-Process Firefox by mixmastamyk on Tuesday August 02, 2016 @10:36PM (#12215181)
Moz makes a few hundred million a year, should be enough to develop a browser, no? At a certain point enlarging a dev team becomes counter-productive.
Re:What’s Next for Multi-Process Firefox by Sylos on Wednesday August 03, 2016 @03:02AM (#12216151)
Well, Google, Microsoft and Apple make a few billions a year. Undoubtedly, those three invest a much lower percentage of that into their respective browsers than Mozilla does, but it could still easily be more than Mozilla has.

Also just in general, Google can display an advertisement 24/7 on the world's most popular webpage, and bundle their browser on Chrome OS and Android. Microsoft can bundle their browser with the most popular desktop OS. And Apple can bundle their browser with iOS and OSX, which are both still pretty popular in their space.

Mozilla has none of that, and would have to invest a lot of money to get it. They can afford the occasional billboard and organize the occasional PR event, but that's about it. The entire rest of their user gain depends on them being decisively better than the competition, so that people themselves recommend Firefox over other browsers.

Re:What’s Next for Multi-Process Firefox by wooptoo on Wednesday August 03, 2016 @03:26AM (#12216258)
They have a strong image advantage too.

They are the only company to put the user's rights, privacy and best interests first and foremost.

For years they were the only ones to carry the open source flag in the browser space. They had the Mozilla Suite when other browsers were closed, they came out with Firefox which was a blast at the time, and now Servo.

They don't play to a corporate agenda, but tend to do «what is right».

Re:What’s Next for Multi-Process Firefox by tracker1 on Wednesday August 03, 2016 @02:58AM (#12216138)
While there is merit to what you say, there's also plenty of places where a divide and conquer can happen with appropriate management of the browser as a whole.

You have the core rendering engine, ancillary features to the application, and the shell of the application itself. From the core rendering engine, you have several areas that are subsets, from svg parsing/rendering, parsing/processing of text/xml/html/css, audio/video, almost all the APIs that the browser offers outside rendering can be implemented by different teams (web-audio, websocket, webrtc, etc).

The communication libraries themselves can be broken off too. There is really a lot that can be handled by many developers.

Now if there were say 15+ developers on any one feature, that is probably too many to be very productive. On the flip side, given the push towards rust, there's room for some duplication of effort there even.

Re:What’s Next for Multi-Process Firefox by kilroy123 on Tuesday August 02, 2016 @08:29PM (#12214561)
Sadly I agree. Which is a shame, because in my opinion they _do_ actually leapfrog them both in privacy.
Re:What’s Next for Multi-Process Firefox by HelloImDumb on Tuesday August 02, 2016 @09:22PM (#12214858)
I see your point, but they don't leapfrog Chrome in privacy when the browser is pwned and leaks its entire process state to an attacker.

Any browser is a massive, rich attack surface with bugs, but Chrome has been cutting edge in its sandboxing, privilege separation, and overall security.

True, Chrome is and has always been a Google marketing vehicle, where the user is the product. But has Mozilla always been at the forefront of privacy, even? Something as basic as private browsing, let's see:

April 29, 2005 Safari 2.0 Private Browsing

December 11, 2008 Google Chrome 1.0 Incognito

March 19, 2009 Internet Explorer 8 InPrivate Browsing

June 30, 2009 Mozilla Firefox 3.5 Private Browsing

Apple, Google, and Microsoft all beat Mozilla to the punch.

Re:What’s Next for Multi-Process Firefox by flomo on Tuesday August 02, 2016 @10:44PM (#12215212)
Well, private browsing really functions as mechanism to discourage people from routinely deleting their cookies. (Which, according to a mid-2000s report, was an extremely common user activity.) Firefox always had an option to delete history/cookies on exit.
Re:What’s Next for Multi-Process Firefox by smt88 on Tuesday August 02, 2016 @09:37PM (#12214927)
Firefox had plugins for private browsing since 2008. Only Apple beat them.

One of Chrome's many small victories was including sensible plugins by default, whereas Firefox's best functionality had to be downloaded separately.

Re:What’s Next for Multi-Process Firefox by HelloImDumb on Tuesday August 02, 2016 @09:59PM (#12215028)
What kind of moving goalpost is that? Chrome 1.0 came with Incognito built-in. Safari had it for three years. Because there may have been a plugin for Firefox a few months before Chrome's release, and followed IE in shipping it, it's the superior privacy option?
Re:What’s Next for Multi-Process Firefox by smt88 on Tuesday August 02, 2016 @10:20PM (#12215123)
1) I didn't say Firefox was better for privacy. I was only pointing out that comparing the browsers based on built-in privacy mode isn't apples-to-apples.

2) On a related note, Firefox left out lots of features because there were well-supported, popular plugins that covered the same thing. Whenever a browser gets built-in ad blocking, should we say that all the other browsers are worse privacy options because they chose to leave those as third-party plugins?

Re:What’s Next for Multi-Process Firefox by Flimm on Wednesday August 03, 2016 @06:12AM (#12216894)
On 2, I would say yes. The default experience matters.
Re:What’s Next for Multi-Process Firefox by Narishma on Wednesday August 03, 2016 @10:59AM (#12218382)
> unless google really drops the ball

They did for me recently when they changed the behaviour of the backspace key without a valid option to set it back to the way it was, so I went back to Firefox instead of futzing about with half-broken 3rd party extensions.