Plus, it's worth repeating that Firefox sync works without problem across desktop and mobile. The sync [0] and auth [1] servers are open source, and one could choose to use one or both on his premises.
IIRC, browser synchronization was a thing on Firefox well before it landed on chrome.
For me, the biggest benefit of using Firefox on mobile is access to extensions. Full-featured uBlock Origin and Privacy Badger make mobile browsing a lot less annoying.
Ublock was enough to make slower Firefox on Android worth using because it makes sites so much faster. Once the quantum features start arriving on mobile its going to be impossible for chrome to match.
Firefox Focus is a mobile browser flavor which requires no installation of any extension, and behaves a bit like windows in "Private" mode but with ad blocking by default.
I use Firefox Focus as my default browser for all news/app/etc links except for when I specifically need access to existing sessions and don't want to log everything in again.
Browser sync in FF also does not sell your data, and works on all major OS-es, including obscure linuxes. I also love the «send tab» functionality - need some info on my desktop tomorrow? Send tab, and it’ll be there when I log in!
> IIRC, browser synchronization was a thing on Firefox well before it landed on chrome.
Aye, it was. It was called Mozilla Weave, and it worked with the very early versions of Fennec (the original name of Firefox Mobile).
Also, your data is securely encrypted at Mozilla, and isn't being profiled (if you disable those features in Firefox).
In Firefox 58 (you can run it today on Android by installing Firefox Beta app) you can run PWA (Progressive Web App) which allows you to turn a website into basically an app. Not every website will work though.
Unfortunately I had the same issue on my Android phone - every time I re-opened the app I had to open a new tab in order to browse. I ended up installing and using Firefox Nightly instead (where the issue seems to be resolved).
Have you tried again since 57 was released? Or tried nightly? I've been using Firefox on mobile for a long time, and I remember experiencing similar issues from time to time, but I haven't had these in a while.
FWIW, closing tabs once in a while can also help. Mobiles typically don't have too much RAM.
My primary issue with Firefox on Android is that in portrait mode, the URL bar has enough width for about five characters -- the rest being taken up by the back / forward buttons to the left, a magnifying glass icon, the QR code icon, a microphone icon, the refresh button, the tabs button, and the "three-dots" options menu. Typing a URL into a text-field that can display five characters at a time borders on ludicrous, and I have no idea how anyone thought that was reasonable to ship that as the default UI.
I'm not sure how your Firefox ended up like that. Mine has the QR button, mic button and X button when entering a URL, or the reader button, tab button and menu button when viewing a page. The back/forward and refresh buttons are inside the menu ('three dots').
>IIRC, browser synchronization was a thing on Firefox well before it landed on chrome.
Yes, but it was bad. Chrome allowed you to sync just by typing your Google account credentials. Firefox relied on sharing a text file between computers. If you lost the file and reset your profile (as was common in old Firefox), you effectively lost all of your synced content. Adding a new computer was difficult and slow. Firefox ultimately failed the "could you explain this to your parents" test, and many users simply gave up trying to get it to work.
I'm about to leave Firefox on Android behind after running it for years (and even running Nighlies/Betas), because they broke the "Share Link To Bottom" plugin and the browser is literally - no exaggeration - causing me to curse and swear every other time I open a link.
I try to stick to Firefox on my desktop and find workarounds for broken addons, but on a mobile it seems easier to switch away.
Massive memory leaks for me, Nexus 5x. Took me a while to realize it was FF basically bricking my phone. Since switching back to Chrome, I've had no issues.
No, because people don't switch to the best product, that switch away from the worst.
People went to Firefox because it was head and shoulders better than IE6 for regular use - popup blocking and tabbed browsing were the most obvious examples, but even then (and being advertised on Google's front page), it only for about 35% of the market.
Now let's say it'll be a bit faster than Chrome. Let's say it'll be twice faster than Chrome, with better deep extensions coming back, it won't make a difference. Chrome is too usable.
I switched to the new Firefox and removed chrome because I tried it and liked it better, not because Chrome is the worst. Also I'm curious how people went from Firefox to Chrome in the first place using your "move away from the worst" logic. Certainly something else happened there or FF was much worst than I remember.
I think google did a «good job» (not as in «good guy» but as in «effectively») pushing chrome by shoving it down your throat on the google search page. They will keep doing it. Therefore, it’ll be the first alternative after booting into windows the first time. As people switch computers, more and more users drift over to chrome. Most people stop bothering at that point, they set up their computers once and consider themselves done.
But most people aren't even going to try Firefox. Unless Google do something seriously stupid with Chrome, the average user has no incentive to move. Chrome is good enough.
I don't know, I feel like many people started using Chrome because technical-minded folk recommended it to them (certainly, I told many family members to switch to it). So, maybe the same thing will happen with Firefox (I have just switched from Chrome to Firefox and am prepared to recommend it as someone's default browser).
Looking towards technical minded relatives and friends definitely helped, but what really pushed Chrome over the top IMO was Google.com, Gmail, Youtube, Google Docs, Google Calendar, Hangouts, etc. all insisting you must use Chrome.
To explain why I think this, the Quality delta between Firefox and IE when Firefox reached about 20-30% marketshare was orders of magnitude greater than the delta between Chrome and Firefox over the period Chrome managed to push to 60-70% marketshare.
I wonder why they did - FF was always safer in so many ways, helping you instead of the advertisers, having a separate certificate store made it safe against nasty BIOS level viruses, etc etc
> having a separate certificate store made it safe against nasty BIOS level viruses
Sorry but that’s just wishful thinking. In this regard Firefox is less of a target than Chrome/Internet Explorer because it isn’t the dominant monoculture, similar to how Mac OS is less of a target than Windows. That doesn’t mean its safe.
Firefox installations quickly got slow and bad. Profiles became corrupt. Third party software installed toolbars that were difficult (or impossible, in some cases) to remove. It wasn't until fairly recently that Mozilla stopped third parties from sideloading crapware extensions. Firefox dev tools got slow to the point of unusability and lacked critical features (e.g., almost anything related to websockets). Standards support lagged far behind.
When Gary Kovacs shifted the focus to Firefox OS, the desktop browser lost pretty much any advantage it once had.
I moved away from Firefox because they removed the Ctrl+E to quick search many years ago. I know Ctrl+E is back on Firefox now but back then there was no way to set my own key commands so I tried Chrome (which kept Ctrl+E) and never turned back.
I stopped using firefox since they are delivering cliqz with every 100th download which will analyze and store your browsing history and is a shame for the mozilla foundation.
I'm using waterfox now btw...
I'm shocked; how is this at all consistent with Mozilla's principles? Situations like this one damage trust; what else is going on that I didn't happen to stumble across? From the announcement:
Users who receive a version of Firefox with Cliqz will have their browsing activity sent to Cliqz servers, including the URLs of pages they visit. Cliqz uses several techniques to attempt to remove sensitive information from this browsing data before it is sent from Firefox. Cliqz does not build browsing profiles for individual users and discards the user’s IP address once the data is collected. Cliqz’s code is available for public review and a description of these techniques can be found here.
And it's opt-out; it's not clear if the users receive an opt-out notification:
Users will receive a version of Firefox with Cliqz recommendations enabled out of the box.
> Do users who receive browsers with Cliqz get a clear, effective opt-out warning? Really, it should be opt-in
Either way would render the experiment useless (self-selection bias). Obviously if something like this should make it into Firefox proper, that would make something like an opt-out warning a hard requirement.
Either way would render the experiment useless (self-selection bias).
But that is Mozilla's problem, not their users'.
As probably would be the legal position in many places, if news of this got out more widely and regulators started getting involved. It's hard to see how this wouldn't fall foul of even general EU data protection laws today, never mind anything stronger that is specific to Germany.
> But that is Mozilla's problem, not their users'.
Sure, I'm just telling you how we arrived at the current situation and why your proposed alternative is a nonstarter for an experiment that tries to figure out if this feature adds user value in the first place.
If the warning is a hard requirement after the experiment, running an experiment without the opt-out warning biases your results. Having the warning in the experiment gives you a better picture of what happens to users when you present them the option.
Except that the question that the experiment wants to answer isn't what happens when you present users with the option. That would only become relevant later. The question right now is whether Cliqz' address bar autocomplete results are superior to the status quo.
It would have to be opt-in or the user would have to be clearly informed, if it actually collected personally identifiable information. That's German law. But it doesn't.
It will not store "your browsing history". It will store the webpages that you've browsed to, without identifying information about you associated with it and the entries will be intermixed with all other users' browsing history entries, so it's not going to be possible to identify you.
I find the Cliqz issue horrifying and am considering Waterfox for similar reasons, but I have to point out that "every 100th download" is a gross misrepresentation - it's an experiment to be tried in "<1% of German users of Firefox" (which is still bad, but a very different thing from every 100th download). According to [this comment](https://www.reddit.com/r/firefox/comments/74n0b2/mozilla_shi...), "the experiment can be removed like any addon."
I still find this a very bad sign about the directions they're willing to explore, but it's important to get the accurate scale of the thing.
One step forward two steps backwards. Why does every large organization on the internet have to be a shithead in 2017. Mozilla - no amount of sanitization will prevent shitty websites sending sensitive data in URLs therefore you don't include this in your product. Even if you obtain consent - because it isn't in your users best interest to have this enabled. I'm extremely disappointed.
Does the new Firefox have a password sync feature like Chrome's? I'd much rather trust Mozilla with that sort of thing (especially if I could host my own backend to it somewhere) than Google.
As an aside, one silly thing that tarnished Mozilla's brand for me was the treatment of Brendan Eich. I may not agree with his donation of money but I think impinging on his rights is a far more fundamental immorality than what he was accused of.
It's like Twitter's controversy over blue ticks. If they'd only given them as a result of identity verification, and not used them to indicate Twitter's approval of someone's beliefs, then they wouldn't be where they are now.
Stick to good tech and fighting for relevant causes, e.g. net neutrality, I say.
Eich did not have his rights impinged on. He chose to resign--he chose that--because the community far beyond Mozilla didn't approve of his donations and he saw that he'd lost a lot of people's trust in him. His actions had consequences, which he recognized and owned.
He has, presumably, gone on to find other work, after resigning, he still has a platform, he can still talk and associate freely.
You can dislike the fact that he resigned, but blaming anyone for his actions but him doesn't make sense.
>Eich did not have his rights impinged on. He chose to resign--he chose that--because the community far beyond Mozilla didn't approve of his donations and he saw that he'd lost a lot of people's trust in him. His actions had consequences, which he recognized and owned.
Yes, this is a key point people forget. No one fired Eich. However, a large chunk of Mozilla's open source contributors (along with a decent chunk of their paid employees) were ready to resign.
Eich may not have had good politics, but he legitimately cared about the organization and was not going to let it implode on his account.
Funny how so-called-libertarians love at will employment until it's used to pressure a bigot to resign.
> However, a large chunk of Mozilla's open source contributors (along with a decent chunk of their paid employees) were ready to resign.
Citation, please. I'm a Mozilla employee, and was a Mozilla employee at the time. I'm not going to claim everyone was ecstatic about things; there was a lot of soul-searching going on for some people. But I don't recall any "Mozilla employees", in the sense of people working for the Mozilla Corporation, saying they would resign, nor calling for Brendan to do so.
There _were_ a few such calls from interns, and a few employees, at the Mozilla Foundation. People who were not involved in Firefox in any way, did not know Brendan, and were not relevant to the things that Mozilla is actually known for doing.
Of course the press spun the story in such a way as to make it seem like there was mass unhappiness internally... And most people read the press spin, not the actual blog posts actual Mozilla employees were writing at the time.
>Citation, please. I'm a Mozilla employee, and was a Mozilla employee at the time. I'm not going to claim everyone was ecstatic about things; there was a lot of soul-searching going on for some people. But I don't recall any "Mozilla employees", in the sense of people working for the Mozilla Corporation, saying they would resign, nor calling for Brendan to do so.
I'm not saying there was an organized "we, a group, will resing". I'm only repeating what I was told hanging out on the deck during one of the cantinas - that there would have be a lot of turnover if he hadn't left :)
IIRC Eich indirectly acknowledged this when he resigned when he said he didn't want to hurt Mozilla.
No, I did not acknowledge that. Stop putting your preferred narrow interpretation on a (too-broad) fact pattern and then attributing your interpretation to me. Doing so takes a low road and can be used against your views as well as for them.
I was merely pointing out that it is my understanding that you resigned, that you mentioned care for Mozilla as why you did, and that many people would have resigned if you hadn't. I thought I made that clear when I said "indirectly acknowledged". I guess my understanding was wrong. Sorry for that.
I definitely don't want to take a low road. We never met, but even the people who disagreed with your politics all thought you were a pleasant person and would mention missing you at the various company events. I hope you know that.
"many people would have resigned" is unfalsifiable, but if you mean the six Mozilla Foundation employees who tweeted on day 2 that I should resign, those people never worked for me. For whatever reason they almost all resigned for other jobs within a year of my leaving, anyway. I hope you weren't thinking of them when making the "would have resigned" imponderable claim.
> Does the new Firefox have a password sync feature like Chrome's?
It's had that for years, and I'm yet to have a problem with it.
> As an aside, one silly thing that tarnished Mozilla's brand for me was the treatment of Brendan Eich. I may not agree with his donation of money but I think impinging on his rights is a far more fundamental immorality than what he was accused of.
I can't decide if what they did was wrong or not, but I'm not totally comfortable with that issue either... no person or company is perfect though.
Does it still allow you to see the password in the settings? I think that’s a not so great idea. I remember messing with friends by getting their saved passwords.
Yes. The same as Chrome. As it should be. If they weren't in the UI, you could just extract them from the browser's data files or process memory. All that preventing access does is annoy users and give them a false sense of security.
The thing with Brendan happened ~3 years ago now. A lot of the leadership that was there at the time is gone now. How long should Mozilla be punished for that?
No one is being punished - I'm not their parent not their government.
Which is sort of my original concern. Mozilla was neither of those things to Brendan Eich, and yet more and more we see employers and influencers behaving as though they are.
I don't mean to rehash old arguments but in his position he was representing Mozilla. And Mozilla is an advocacy organisation with a deep interest in policy. They aren't just a browser maker; Firefox is merely their most widely known campaign.
If the organisation felt he didn't represent them then it doesn't make sense for him to be CEO. It would have been less of an issue if he were in a more purely technical role.
That's my point - I don't want them to spend their money on being a general-purpose morality organisation.
I want a browser maker who specifically argues for the most foundational and important of needs regarding the internet. I don't want a social justice crusader who values PR above all.
Ah, the acceptable bigotry of "those [people in my outgroup] can be CTOs but not CEOs". Well done.
In fact, my leaving had nothing to do with advocacy _per se_. As Erin Kissane wrote at the time, if Mozilla were to become a narrow "progressive" advocacy group, it would fail at its stated mission. Whether it has taken or will take that fall, I won't attempt to evaluate here. But that's the risk Erin cited, and I agree.
> As an aside, one silly thing that tarnished Mozilla's brand for me was the treatment of Brendan Eich. I may not agree with his donation of money but I think impinging on his rights is a far more fundamental immorality than what he was accused of.
Well, and Chrome's maker fired someone for a certain inflammatory manifesto. I don't think that at this point it is realistic to run a tech company of any sort in the US without some amount of displays of political loyalty to the vocal majority of your recruitment and customer base.
"As an aside, one silly thing that tarnished Mozilla's brand for me was the treatment of Brendan Eich. I may not agree with his donation of money but I think impinging on his rights is a far more fundamental immorality than what he was accused of."
It wasn't about his rights as a private person. It was that he gave a non-trivial amount of financial support to a campaign to remove rights from other people, which did not seem consistent with the values of the non-profit that he wanted to publicly represent as CEO.
The privacy-conscious / anti-Google crowd isn't large enough to move the needle. For Firefox to regain market share, one of two things must happen:
1) Google makes a really poor decision (e.g. disabling ad blockers, charging $$)
2) Mozilla partners with existing web apps or they build their own web apps with extra value that is only available in Firefox (in the same way that Google sites generally perform better on Chrome than FF)
What evidence do you have of this? As far as I can tell, their strategy has been to 1) imitate Chrome and 2) get faster. I don't know that people wanted them to imitate Chrome and get faster, and I would cite the fact that millions have uninstalled it over that period, and installed Chrome.
The “containers” thing they’re doing to have separate containers with separate cookie jars, is something I’m loving that some die hard Chrome fans have said “I want that” when I explain it to them.
It allows me to have a work container logged in to work trello and Gmail, and a personal container logged in to personal trello and Gmail. It also allows me to have a FB container so I can stay logged in there, but do all my browsing in other containers so they don’t know which other sites I’ve been visiting. That’s a feature I didn’t know I wanted until they built it.
I totally believe that they could attract a few extremely technically savvy people, but not enough of them quickly enough to be more than insignificant. I mention "quickly" because if it turns out to be a significant attraction, Chrome could just implement it.
I remember when Chrome has started their project. People were mainly switching because: tabs crashed independently from the browser, and it was way faster than Firefox. You don't need much to get people to switch.
At the beginning, yes, and the process separation was heavily marketed so people were switching for that reason even if there was actually no material benefit that would show up in their own typical usage. But on the other side, those early switchers had no reason to stay with firefox because they weren't heavy add-on users, and both browsers were nominally open source and anti-IE.
I think the bleeding of users from Firefox over time, however, has been because of a Chrome inspired cycle of constant add-on breakages, removals of configuration options, and UI changes towards a Chrome look, with each breakage and change removing a distinction between the two browsers that would keep a user from switching. Additionally as the add-on breakages were happening, Chrome was expanding their library of add-ons, further removing distinction between the two browsers, and in some cases, such as developer tools, gaining the upper hand.
I think it's important to note that this latest firefox move was not only a move to render it incompatable with its own library of add-ons, but to adopt Chrome's library as its own. The multiprocess functionality that it is now touting was a response to Chrome's own multiprocess functionality. The UI is now completely indistinct from Chrome's.
For every add-on that didn't make it over, there's 1000 users for whom that add-on was keeping them from switching to Chrome. All of the UI manipulating add-ons represented ingrained habits that 1000 users really didn't want to change, so they were bound to firefox. At the point of this version change, upgrading firefox or moving to Chrome has exactly the same amount of friction. The only way Firefox can compete with Chrome now that they're indistinguishable is on marketing, and I'm fairly sure that Google is the biggest advertising company in the world.
I suspect that this has been intentional. Firefox is again financially dependent on Google. This was as precipitous a fall as the Elop-headed Nokia.
They can still compete on privacy (Chrome won't do that), aggressive anti-ad stance (same), future improvements in security (if they manage to move more elements like JS runtime to Rust and improve sandboxing).
For the first one they need to stop undermining their own message though (by adding weird extensions people don't want). For the last one, it will take years to complete and some more time for the trend to be noticed.
Banking on web apps to save Firefox won't work out while the browser continues to not support CSS styling of scrollbars. It's so fundamental to so many modern web apps- and a major part of why they tend to feel slightly worse on Firefox.
Non-tech folks don't care about rendering engine. They might fall for tons of advertisment that Chrome had or they might be force-fed a new browser with a new computer or OS purchase, none of these things are going to happen for FF.
Tech folks might care about shiny new engine, but they are also more likely to care about their favorite extensions not working anymore. I personally wouldn't be surprised if FF actually lost some share in the long run, but that's just my dissapointment talking.
I know for certain that if something like waterfox keeps old extensions running I will gladly give up Quantum for a better UX.
My experience is anecdotal of course, but IMHO "fast" is something tech savy users are looking for. "Which one is the Internet" or "I know this icon from a big shiny ad" is how non-tech folks function.
Full disclaimer: Years spent supporting end users might have negatively affected my view of the "average user".
Yeah, I do think your view is a bit skewed. There's an ever-growing number of users that grew up with computers, who know at least how to install a different program on their own, despite most definitely not being a techie either.
And even among techies, I think you overestimate how many actually care about much else than "fast" when it comes to a browser. Just look at how many techies run Chrome. It's usually not the most wizardly of techies, but a good number of techies either way.
I have been using Firefox again for a while now after completing my Google exit and I’m enjoying every single day with Quantum. The real game changer for me was the introduction of container though. I only wish they also had separated history.
The question isn't whether the new Firefox is better than older versions, but rather whether they can regain any serious market share. And for that to happen, it won't be enough for Firefox to be comparable or even slightly better than Chrome.
Yes, really. Why should that surprise you? The majority of XUL extensions had not been ported to WebExtensions when Mozilla pushed this update and Mozilla was well aware of that fact. This was executed poorly and Mozilla deserves to be criticized for it.
XUL extensions e.g. the user interface could not be ported, the notice they were given was "this will stop your extension working" with users being given the advice from advocates "don't worry we'll build some replacement APIs sometime soon..."
There were years of notice, but no infrastructure to support addons migrating to WebExtensions. They're still working on the supporting APIs now, after throwing the switch. What would have made more sense would be to bring in support for WebExtensions and all APIs necessary to replicate popular addons first, give them enough time to migrate, and then disable legacy extensions.
Technically yes, there is the ESR version available until April, but the vast majority of regular users aren't going to know about that.
By far the vast majority of popular extensions were ported well in advance, NoScript[1] was a notable exception, but even Tree Style Tabs is available now.
[1] I've found uBlock Origin in dynamic mode to be superior to NoScript, anyway.
TST is available, but in its default state it has tabs listed across the top and also down the side. You have to edit some other file manually in order to hide the tabs from appearing at the top. TST tells users about this but also disclaims the possibility that following the instructions to do so may break things. This is not exactly a smooth transition.
Right on, was using hide the title bar[0] add-on which gives you a script to install, and then it runs the script on window creation, but need the keymap since it comes back after every window resize. Just need to edit the extension to run on resize really, if possible.
Well, large parts of Chrome's market share is caused by their aggressive marketing, which is likely quite effective with casual users. And those probably won't switch for performance or privacy reasons.
No, no chance Firefox could get back to owning 20% or 30% again.
Just look around you, how many of the people you know made the switch? Most of the people I know gave the new FF a try, concluded that it's fine and got back to Chrome. Right now, there's just no reason to switch.
Also, everybody's going about FF consuming less memory than Chrome, but in my short personal test there was no real difference.
Long time Firefox user every since I left safari I tried Firefox and chrome and didn't like chrome at all. Now I only run some linux with Firefox but have chrome only as a back up to test or if YouTube has a glitchy anti Firefox day.
I think currently, market share isn't dictated by quality, but more by pushing stuff on users (Google is really obnoxious with it).
So it's not as much of a technical problem, as a marketing one.
May be mandating browser choice for mobile devices could help it, including forcing Apple to remove the ban on competing browsers, which really should have been tackled by anti-trust a long time ago.
After my first few hours using desktop Quantum I thought for sure I was switching away from Chrome. But then videos stopped playing (audio would play and the video would just be black) and didn't work again until restarting the browser. And then it just kept happening every time it had been open for a couple hours. Back to Chrome for now.
Did you report the problem? Given how many video-related problems end up being caused by drivers that seems like it'd be a potential entry on the blacklist.
I tried it for a week hoping to reduce the amount of swapping my laptop did during a normal workday. It wasn’t any better than chrome unfortunately and due to the lack of workaround extensions like the Great Suspender it was actually worse.
A couple of hours ago, I've had about 25 tabs opened, about 20 of which were suspended. Just for the heck of it, I've checked the memory usage of Firefox. 500 megabytes.
I'm in the same boat, but the progress they have been making as of late is really impressive, and it was markedly better than the last time I tried it.
Will probably try again in another 6-8 months. I have problems with Chrome eating tons of memory, hoping the new FF gets a lot leaner (and robust) as their newer tech gets integrated. Plus, the blog posts on their usage of Rust are really fun to read!
The new version of Firefox is great. And a massive improvement over previous versions.
However I think Mozilla have a lot of work to do. Desktop browsing is pretty much legacy at this point so convincing mobile users to use something other than the default browser will be hard.
I work on a site with 8 figure monthly unique visits and the browser stats are startling.
iOS Safari had about 51%, Chrome mobile & desktop about 35%, Samsung browser at 7%. Nothing else has over 2-3%. Edge, for example, is a rounding error.
I used to work with a developer who'd preach the "IceWeasel is the ultimate browser" gospel to us. I guess this makes your argument somehow more valid:)
OS might be hard, as Win and Mac have their own browsers, so my first guess would be Linux or security focused builds of Android. This doesn't sound like enough exposure to make a significant/disruptive change.
I expect Mozilla to be more active in this area though. Yes, performance is important and I switched to FF recently too, but I'm still using Electron powered apps on my MacBook.
What's more plausible is a situation where Chrome for some reason (ekhm... privacy?) loses market share to FF/IE/Safari. Having more competition will make my browser better, regardless of my choice.
Firefox is the default browser in virtually all distros, even Debian.
As for Android, to make a difference it would need to be installed in Gapps-by-default phones, so Google would have to concede. Maybe if the EU hits them with an antitrust lawsuit Microsoft-style, otherwise it's not gonna happen.
I'm in a similar boat waiting for decent alternatives to 'Speed Dial' and 'FireGestures' addons. The existing 'VimFx' replacements like Saka Key are also only partially there in terms of features.
I'm wary of switching to Waterfox because it's a single developer project and might be more buggy than it seems from the outside, but I might end up there after all, if equivalent addons don't rise up.
For anyone that thought that Microsoft was changing its ways, that should be a big warning:
> Even worse, the new Windows 10 S operating system prohibits the installation of any apps which do not come from the Windows Store, so it is impossible to install Firefox on devices running Windows 10 S.
Everyone keeps talking about how Firefox is faster than Chrome, but for me Chrome is night and day faster at rendering. And that's comparing Chrome with about a dozen extensions vs a fresh install of Quantum. And yes, that's without cache
This morning I came into work to a sluggish FF browser. One of the FF processes was using 1.6GB of memory, and one entire core of CPU. I had 1 pinned and 5 normal tabs open. I'm not particularly impressed with the reality of the new FF.
No. Most of today's users are using mobiles to surf the web. The new Firefox version is a gamechanger on desktop, but on par with Chrome and other mobile browsers.
For me, Quantum is the first usable Firefox version on mobile. I've been trialing various versions of it over the years in desperate hope that it would get to a usable state, because the mobile browser ecosystem is bleak other than Firefox - it's the only browser with actual features I want. I was only held back from it by performance.
So in that sense, I'd say it's a complete gamechanger on mobile, much more than on desktop.
Yeah but it's still noticably more usable because of the UI change in this release. It basically looks more like Chrome for Android; where as before it looked like a Android Cupcake app.
BUT Quantum is also coming to mobile. Nevertheless, i've already switched to Firefox on desktop and mobile and i'm extremely happy with it after years of Chromium usage.
Maybe on Android, but it's not possible to run a different browser on iOS. Well, you can install another browser, but it is really just a different interface for the Safari browser. The guts of the browser will still be Safari.
It hasn't stopped either Chrome or Firefox[0] from releasing Safari skins on the AppStore, but unfortunately neither are actually Chrome[1] or Firefox[2]
The biggest selling point for me with firefox on android is the ability to use an adblocker. There's no way I'm going to consider using chrome on android unless it supports extensions like ublock and whatnot.
UBlock Origin in dynamic mode is absolutely essential, I will not use a browser where it is not available. So that means Firefox on mobile, and it makes browsing much less aggravating.
> Even worse is Google’s policy of prohibiting the installation of 3rd party web browsers on Chrome OS devices. Firefox simply can’t be installed on any Chromebook.
The Android version of Firefox can be installed on Chrome OS though.
It can, but Quantum was basically just the very start of that journey. They'll need to try and keep one step ahead of Chrome with every release. If they can do that, they'll rebuild the following.
Generally, the new FF is great. One small criticism - their GPU accelerated CSS animations are a bit choppy on mobiles (these a generally very smooth on chrome and Safari).
I tried it on Android, but a few of my app's users complained and a lot of them use iOS. You can try it out at https://usebx.com/app - just click try demo to get to where the animation is.
You'll find the page transitions are not quite as smooth in Firefox when compared with Chrome or Safari.
Well, as it says in the article, Firefox and Chrome on iOS are both just WebKit wrappers, because Apple doesn't allow other browser engines into their store, so I don't know what kind of black magic or bias or whatnot your users are experiencing.
As for Android, the new CSS engine (called "Quantum CSS" or "Stylo"), which landed with Firefox 57 on the desktop, should ship on Android with Firefox 59.
There's also "Quantum Render" or "WebRender" upcoming somewhen soonish. I'm guessing Firefox 60 or 61 on the desktop, so maybe again with a small delay then on Android, I don't know.
Also, noticed another thing - on Android, using FF nightly, I couldn't access a local network URL via FF (Unable to connect message), but could with Chrome - example URL http://10.0.130.64/path
Firefox' path to market share growth has to go through mobile. How to get users to switch from preinstalled good-enough Chrome is the tough nut to crack, though.
I wonder if the EU will start to consider the situations with iOS and Android equivalent to Microsoft bundling IE ... giving users an obvious post-install choice might help a lot.
Google Inbox web app is laggy af in Chrome for me, honestly. I use it for work. I'm on Firefox Nightly and it's super fast everywhere except for the Google Web Apps (which aren't butter smooth on any browser, even Chrome).
That's funny because gmail on Quantum was actually the final straw for me. I was liking what I was seeing trying Quantum out and then I loaded up gmail and boom it worked and looked better than gmail on Chrome and I was sold.
Quantum was annoying enough to kill Firefox for me completely.
I've completely shifted.
EDIT: Added explanation
The big issue is I often switch between Browsers for work.
So I often switch between Chrome, Firefox, and Opera with custom extensions for my workflow.
With the amount of choice, I don't need to stick with a browser that makes my life harder than easier.
Chrome is my main browser now, it was already getting close to majority usage for me.
I feel like there's a lot of anti-Mozilla trolls on this particular comments section. Compared to most other articles on HN, it just doesn't align with the views I've heard elsewhere, and certainly doesn't align with my experience as a happy Firefox user.
How do you tell "anti-Mozilla troll" from a genuinely dissapointed user? I'm currently on FF ESR, once my favorite extentsions stop working there I might as well switch the browser completely, since my UX will be completely reset just as it would be with a entirely different browser.
I'm not saying I will switch to something else (other browsers seem equally unusable to me as FF57 is currently), but I will certainly try all the options to find out if something else isn't at least marginally better than FF57.
I have been using Firefox for about 5 years, I'm still using it and despite that I'm anti-Mozilla. The browser performance has always been fine to me and 57 doesn't look particularly "faster" as they are advertising it.
I'm one of the dozen people that care about a browser being user configurable and not doing any analytics, telemetry, experiments,... . Mozilla has lately been working exactly against this. I have about 80KB of patch files just to remove all the crap Mozilla managed to stuff into an otherwise perfect browser like: advertisements, telemetry, more data sharing, geolocation using google servers, a social network and proprietary binary blobs for service integration.
When I build a clean version of Firefox then I'm happy too.
Simply put yes.
The big issue is I often switch between Browsers for work.
So I often switch between Chrome, Firefox, and Opera with custom extensions for my workflow.
With the amount of choice, I don't need to stick with a browser that makes my life harder than easier.
Chrome is my main browser now, it was already getting close to majority usage for me.
I don't understand your comment. Are you saying that Chrome has extensions that Firefox doesnt? Or are you saying that you used Firefox because of certain extensions it had and other browsers didnt, but Chrome was still your primary browser, and that now the FF extensions you used don't work anymore, you don't see any reason to use FF anymore?
Extensions you have for Chrome can work on Firefox with little to no modification. The extensions you rely on probably just need to be repacked into a WebExtension.
Plus, it's worth repeating that Firefox sync works without problem across desktop and mobile. The sync [0] and auth [1] servers are open source, and one could choose to use one or both on his premises.
IIRC, browser synchronization was a thing on Firefox well before it landed on chrome.
[0] https://mozilla-services.readthedocs.io/en/latest/howtos/run...
[1] https://mozilla-services.readthedocs.io/en/latest/howtos/run...